Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 3072 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 4.9980. Min Range: -4.9980

Max Act: 4.9980. Min Act: -0.0000

Data Index: 834839 (C4 (Web Text))

Max Activating Token Index: 1003

Text #1

Max Range: 4.9980. Min Range: -4.9980

Max Act: 3.8283. Min Act: -0.0000

Data Index: 1258306 (C4 (Web Text))

Max Activating Token Index: 435

Text #2

Max Range: 4.9980. Min Range: -4.9980

Max Act: 3.1721. Min Act: -0.0000

Data Index: 920944 (C4 (Web Text))

Max Activating Token Index: 730

Text #3

Max Range: 4.9980. Min Range: -4.9980

Max Act: 3.2525. Min Act: -0.0000

Data Index: 368834 (C4 (Web Text))

Max Activating Token Index: 177

Text #4

Max Range: 4.9980. Min Range: -4.9980

Max Act: 3.0372. Min Act: -0.0000

Data Index: 477131 (C4 (Web Text))

Max Activating Token Index: 79

Text #5

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.4745. Min Act: -0.0000

Data Index: 745252 (C4 (Web Text))

Max Activating Token Index: 313

Text #6

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.7801. Min Act: -0.0000

Data Index: 1040749 (C4 (Web Text))

Max Activating Token Index: 151

Text #7

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.7615. Min Act: -0.0000

Data Index: 180624 (C4 (Web Text))

Max Activating Token Index: 686

Text #8

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.6129. Min Act: -0.0000

Data Index: 222742 (C4 (Web Text))

Max Activating Token Index: 221

Text #9

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.5645. Min Act: -0.0000

Data Index: 917329 (C4 (Web Text))

Max Activating Token Index: 749

Text #10

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.5379. Min Act: -0.0000

Data Index: 642983 (C4 (Web Text))

Max Activating Token Index: 507

Text #11

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.5915. Min Act: -0.0000

Data Index: 539225 (C4 (Web Text))

Max Activating Token Index: 635

Text #12

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.4568. Min Act: -0.0000

Data Index: 1282233 (C4 (Web Text))

Max Activating Token Index: 203

Text #13

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.3284. Min Act: -0.0000

Data Index: 909158 (C4 (Web Text))

Max Activating Token Index: 241

Text #14

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.4363. Min Act: -0.0000

Data Index: 284493 (C4 (Web Text))

Max Activating Token Index: 298

Text #15

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.4387. Min Act: -0.0000

Data Index: 659678 (C4 (Web Text))

Max Activating Token Index: 151

Text #16

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.4188. Min Act: -0.0000

Data Index: 1187064 (C4 (Web Text))

Max Activating Token Index: 929

Text #17

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.4049. Min Act: -0.0000

Data Index: 244987 (C4 (Web Text))

Max Activating Token Index: 343

Text #18

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.3817. Min Act: -0.0000

Data Index: 1298522 (C4 (Web Text))

Max Activating Token Index: 86

Text #19

Max Range: 4.9980. Min Range: -4.9980

Max Act: 2.3694. Min Act: -0.0000

Data Index: 55563 (C4 (Web Text))

Max Activating Token Index: 393